The prognosis for most patients with primary anaplastic CNS tumor, the glioblastoma, is 8-12 months after diagnosis. The outlook is somewhat better for less common tumors, such as anaplastic astrocytoma with a 38-50% 2-year survival, but most anaplastic CNS tumors are highly resistant to currently available therapy. Occasional responses to chemotherapy are seen in recurrent tumors, but these responses are generally of short duration, and cures are rare. Median survival of patients with cancer metastatic to the brain, when untreated, is approximately four weeks from the time of diagnosis. Surgical and radiotherapeutic intervention increase survival, although the outcome is still grim with median survival less than one year. First recognized in 1870 neoplastic meningitis is now being seen with increasing frequency, no doubt reflecting more effective therapy of systemic cancer as well as heightened awareness nd improvement in diagnosis. Current therapy of leptomeningeal disease is particularly ineffective with external beam radiotherapy and intrathecal chemotherapy, specifically methotrexate, thiotepa, or cytosine arabinoside only providing modest benefits, with mean survival following leptomeningeal tumor spread measured in months. Newer therapies are needed for treatment of patients with neoplastic meningitis. The hypothesis of this proposal is that regional therapy combined with systemic therapy of primary brain tumors, metastatic brain tumors and neoplastic meningitis with radiolabeled monoclonal antibodies (MAbs) or bifunctional alkylating agents can substantially enhance therapy of these fulminant malignancies. The specific aims of this proposal are 1) to define the toxicity and activity of intrathecal radiolabelled MAb (murine and chimeric) 81C6, MAb fragment Mel-14 (murine and chimeric), and new MAbs, in the treatment of patients with neoplastic meningitis; 2) to define the toxicity and activity of intracystic radiolabelled MAb (murine and chimeric) 81C6, and new MAbs, in the treatment of patients with newly diagnosed or recurrent cystic gliomas; 3) to define the toxicity an activity of radiolabelled mAb (murine and chimeric) 81C6, MAb fragment (murine and chimeric) Mel-14, and new MAbs administered via a surgically created cystic resection cavity in the treatment of patients with newly diagnosed or recurrent primary or metastatic malignant brain tumors; 4) to define toxicity and activity of intrathecal melphalan and other alkylating agents in the treatment of patients with neoplastic meningitis; 5) to define the toxicity and activity of arterial 4-hydroperoxycyclophosphamide, melphalan, and other alkylating agents in the treatment of patients with newly diagnosied or recurrent anaplastic gliomas; 6) to define the toxicity and activity of intrathecal monoclonal antibody pseudomonas toxin conjugate B3-Lys PE38 in the treatment of patients with neoplastic meningitis; and 7) to conduct in years 3-5, Phase 2 and 3 studies combining systemic therapy reaching the entire neuraxis with the most promising regional MAb/alkylator therapy evaluated in Specific Aims 1-6.230